CN112423027A - Mobile streaming media edge collaboration distribution device and method based on differential privacy - Google Patents
Mobile streaming media edge collaboration distribution device and method based on differential privacy Download PDFInfo
- Publication number
- CN112423027A CN112423027A CN202011140075.1A CN202011140075A CN112423027A CN 112423027 A CN112423027 A CN 112423027A CN 202011140075 A CN202011140075 A CN 202011140075A CN 112423027 A CN112423027 A CN 112423027A
- Authority
- CN
- China
- Prior art keywords
- streaming media
- user
- cooperative
- relay transmission
- representing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
- H04N21/64723—Monitoring of network processes or resources, e.g. monitoring of network load
- H04N21/64738—Monitoring network characteristics, e.g. bandwidth, congestion level
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/239—Interfacing the upstream path of the transmission network, e.g. prioritizing client content requests
- H04N21/2393—Interfacing the upstream path of the transmission network, e.g. prioritizing client content requests involving handling client requests
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25808—Management of client data
- H04N21/25841—Management of client data involving the geographical location of the client
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/266—Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
- H04N21/2662—Controlling the complexity of the video stream, e.g. by scaling the resolution or bitrate of the video stream based on the client capabilities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
- H04N21/8456—Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Security & Cryptography (AREA)
- Computer Graphics (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention belongs to the technical field of mobile edges, and discloses a mobile streaming media edge cooperative distribution device and a mobile streaming media edge cooperative distribution method based on differential privacy, wherein the mobile streaming media edge cooperative distribution device comprises a data privacy module and a cooperative decision module; when the streaming media cooperation transmission period begins, collecting resource state information of an edge server and request video clip information of a streaming media watching user; acquiring fuzzification request information of a streaming media watching user; establishing a geographic position privacy model; acquiring the cooperation information of a relay transmission user; establishing a cooperation cost privacy model and a reward function decision model; constructing a cooperative user matching model, and determining a streaming media content distribution path according to a reward function decision model; and finishing edge cooperative distribution of the streaming media content and updating the related information. The method and the device solve the problem that the privacy of the sensitive data cannot be guaranteed in the streaming media cooperative transmission in the prior art, can prevent the privacy data of the user from being leaked, and reduce the cost of the streaming media cooperative distribution.
Description
Technical Field
The invention relates to the technical field of mobile edges, in particular to a mobile streaming media edge cooperative distribution device and method based on differential privacy.
Background
With the explosive growth of the number of mobile devices and the rapid development of wireless communication technologies, the global data traffic of mobile networks is growing exponentially. According to cisco visual network index predictions, the global internet traffic will increase nearly 3-fold from 2017 to 2022, where mobile streaming media traffic occupies 60% of the global internet traffic in 2017, which is expected to rise to 78% by 2022. The key to success of mobile streaming media services is to provide high-bit-rate, low-delay and distortion-free streaming media content for users, and the burst-growth streaming media data flow makes the traditional streaming media content distribution mode of accessing the core network through a base station difficult to meet the user requirements, so that the streaming media content distribution mechanism needs to be optimized from both the network architecture and the transmission technology.
In order to improve the storage, calculation and bandwidth resources of the mobile internet in the existing network architecture and achieve fast and stable distribution of streaming media content, the probability of mobile edge computing (mec) is firstly proposed in 2014 by the european Telecommunications Standards institute (etsi), which is defined to provide internet service environment and cloud computing capability in the range of a wireless access network near a mobile device, and has the characteristics of proximity to users, low delay of service processing, low load of a backhaul link, mobile perception of a user side and the like. The streaming media service provider deploys the service in the edge server closer to the mobile user, so that the edge storage and the rapid distribution of the streaming media service are realized, and two network bottlenecks of bandwidth and delay in the distribution process of streaming media content are effectively solved.
Meanwhile, as the demand of users for viewing streaming media increases, the streaming media transmission technology is changing day by day, and the data transmission mode is changed from the traditional real-time transmission protocol RTP based on the user datagram protocol UDP to the hypertext transmission protocol HTTP transmission mode based on the transmission control protocol TCP. Each large streaming media service provider continuously provides a new streaming media transmission technology for acquiring more users, and the most widely used technology at present is a code stream adaptive has (http adaptive streaming) technology combining the coding characteristics of mobile streaming media, which encodes a video file into different resolutions and segments the video file into a plurality of segments for transmission, so as to realize dynamic switching of streaming media among different resolution segments and provide high-quality streaming media service. Further considering that the experience quality of streaming media users in a wireless network not only changes along with network conditions such as channel attenuation of the streaming media users, but also is influenced by factors such as channel interference of other users, network competition and the like, streaming media service providers provide a multi-user code stream adaptive HAS-based multimedia cooperative distribution technology which can be applied to mobile edge computing, and a basic framework is composed of three parts, namely streaming media watching users, relay transmission users and an edge cooperative platform. The streaming media watching users are taken as receivers of streaming media contents, and two content distribution paths of streaming media direct transmission and collaborative transmission exist. The direct transmission is directly sent to a target streaming media watching user by an edge server through a wireless link; and the cooperative transmission is carried out by other intelligent terminal users as relay transmission users, the video file is obtained from the edge server, and then the video file is transmitted to the streaming media watching user through D2D communication. D2D communication is a technology that allows end users to reuse radio spectrum resources for direct communication under the control of a cellular system, and has the advantages of stable transmission, reliable interference, and automatic connection. Therefore, the streaming media content cooperative distribution mechanism can avoid the reduction of the experience quality of part of streaming media users caused by user competition and channel attenuation, and effectively ensure the stable and rapid distribution of the streaming media content.
However, the existing mobile streaming media edge collaborative distribution technology based on multi-user code stream self-adaptation has great limitation in real application, because the technology lacks the problem of privacy data leakage of participating users in the collaborative process. The data transmission rate in the D2D communication network is determined by the transmitting power of the terminal equipment and the relative distance between users, so that the power cost, the code rate and the delay consumed by different relay transmission users for cooperatively transmitting video files in the mobile streaming media edge cooperative distribution technology are different, and in order to maximize the overall benefit of a streaming media cooperative transmission architecture, the key is to select a proper streaming media watching user and a relay transmission user to form a cooperative transmission group, and an edge cooperative platform collects user information to complete the decision of matching the cooperative transmission group. However, the existing streaming media cooperative transmission architecture has a risk of leakage of user private data, and compared with a cloud center, the edge cooperative platform is more vulnerable to external attacks, so that user data are maliciously collected. Given that users are rational and selfish, potential privacy data leakage issues result in the streaming media viewing user and the relay transmission user declining to participate in the assistance transmission architecture. Therefore, based on the assumption that the edge collaboration platform is honest but untrustworthy, that is, the edge collaboration platform can give an honest and effective streaming media collaboration matching decision according to the user data, but cannot guarantee the safety of the user data, and the problem that the streaming media collaboration transmission method which guarantees the privacy of sensitive data and is efficient is still worthy of solving is provided.
Disclosure of Invention
The invention provides a mobile streaming media edge cooperative distribution device and method based on differential privacy, and solves the problem that the streaming media cooperative transmission in the prior art cannot ensure the privacy of sensitive data.
The invention provides a mobile streaming media edge collaborative distribution method based on differential privacy, which comprises the following steps:
step 1, when a streaming media cooperation transmission period begins, collecting resource state information of an edge server and request video clip information of a streaming media watching user;
step 2, acquiring fuzzification request information of a streaming media watching user; establishing a geographic position privacy model;
step 3, acquiring the cooperation information of the relay transmission user; establishing a cooperation cost privacy model;
step 4, establishing a reward function decision model, wherein the reward function decision model comprises a streaming media viewing effect model, a relay transmission benefit model and a cooperative reward decision model;
step 5, constructing a cooperative user matching model, making a code rate selection decision and a relay transmission user matching decision for the streaming media distribution process by adopting a streaming media cooperative transmission user matching method based on a Kuhn-Munkres algorithm according to the reward function decision model, and determining a streaming media content distribution path;
and 6, finishing edge cooperative distribution of the streaming media content, and updating the state information of the streaming media cooperative transmission period and the state information of an edge server, a streaming media watching user and a relay transmission user.
Preferably, in step 1, the resource state information of the edge server includes a current network load, a computing resource usage rate, a bandwidth resource allocation condition, and a channel weakening parameter;
the request video clip information of the streaming media watching user comprises the number of the request video clip, the length of the request video clip and the code rate version set of the request video clip.
Preferably, in the step 2, a specific implementation manner of obtaining the fuzzification request information of the streaming media viewing user is as follows: constructing a fuzzification function based on the real geographical position of the streaming media watching user and the localized differential privacy to obtain the false geographical position of the streaming media watching user; acquiring fuzzification request information of the streaming media watching user based on the false geographical position information of the streaming media watching user and the code rate version set of the request video clip;
the real request information of the streaming media watching user is represented as:
Dn,r＝{rn,pn}
wherein D isn,rRepresenting the real request information of the streaming media watching user n in each streaming media cooperation decision period; r isnRepresenting a set of bitrate versions of the requested video segment; p is a radical ofnRepresenting the real geographical location of the streaming media viewing user n;
the fuzzification request information of the streaming media watching user is represented as:
D'n,r＝{rn,p'n}
wherein, D'n,rShowing fuzzification request information of a streaming media watching user n in each streaming media cooperation decision period; p'nRepresenting a false geographical location of the streaming media viewing user n.
Preferably, the geographic location privacy model in step 2 adopts a two-dimensional warping mechanism based on differential privacy, and the corresponding geographic location fuzzification function is represented as:
wherein, O (p'n|pn) A geographic position fuzzification function representing a streaming media viewing user n; p is a radical ofnRepresenting the true geographical location, p, of a streaming media viewing user nn＝{xn,yn}；p'nRepresenting a spurious geographic location, p ', of a streaming media viewing user n'n＝(x'n,y'n) (ii) a Delta represents the information torsion degree in the two-dimensional torsion mechanism, X represents the number of the selectable user horizontal coordinates, and Y represents the number of the selectable user vertical coordinates;
the information torsion degree and the first differential privacy parameter satisfy the following constraint relationship:
ε1＝log(XY-1)+2log(1-δ)-log(δ2-2δ)
wherein epsilon1Representing a first differential privacy parameter.
Preferably, in step 3, the cooperation information of the relay transmission user includes: in each streaming media cooperation decision period, the relay transmission user obtains the rate of the video file from the edge server and the rate of transmitting the video file in D2D communication;
wherein R ism,nRepresents the rate of transmission of video files in D2D communication, B represents the channel bandwidth, PmAnd PcRespectively representing relay transmission of streaming mediaTerminal transmit power, N, for user m and cellular uplink multiplexed user c0Power, σ, representing white Gaussian noisem,nRepresenting the channel gain between a streaming media relay transmission user m and a streaming media viewing user n, dm,nRepresents the distance between a streaming media relay transmission user m and a streaming media viewing user n, mu represents the track attenuation coefficient, sigma0Representing the Gaussian channel coefficient, σn,cRepresenting the channel gain between the stream media relay transmission user m and the channel multiplexing user c, dn,cRepresenting the distance between the streaming media viewing user n and the channel multiplexing user c.
Preferably, in step 3, the collaborative cost privacy model adopts an exponential mechanism based on differential privacy, and a corresponding collaborative cost fuzzification function is represented as:
p (b '| b) represents the probability that the real cooperation cost b is converted into any false cooperation cost b', and fuzzified cooperation cost is output in a random selection mode; q (b' | b) represents a scoring function of the exponential mechanism; Δ b ═ bmax-bminRepresenting the global sensitivity of the scoring function; epsilon2Representing a second differential privacy parameter.
Preferably, in the step 4, the benefit of the streaming media watching user n in the streaming media watching effect model is defined as:
wherein Q isn(D'n,r) The method comprises the steps of (1) representing the user benefit of viewing the streaming media, and theta represents the viewing benefit of the streaming media with unit bit rate; h ism(D'n,r) Is a binary variable, h is the matching success between the video watching user n and the relay transmission user mm(D'n,r) 1, otherwise hm(D'n,r)＝0；
Representing the video rate level,/nRepresents the video segment length, sigma represents the economic loss of the viewing streaming media due to the pause,
code rate for transmitting relay transmission user m to stream media watching user n
Time of consumption, τnRepresenting the time, q, at which the streaming media watching user n plays the video segmentm(D'n,r) The cooperative reward that the streaming media watching user n needs to pay the relay transmission user m is represented; []+Indicates rounding up when
When the temperature of the water is higher than the set temperature,
otherwise
In the relay transmission benefit model, the benefit of the streaming media relay transmission user m is defined as:
wherein Q ism(D'n,r) The user benefit of the relay transmission is shown,
represents the amount of power consumed by the relay transmission user m to download the unit bit video file from the edge server,
representing the amount of power consumed by the relay transmitting user m to download the video file from the edge server per unit time, BmIndicating relay transport user m from the edgeThe rate at which the server obtains the video file,
represents the amount of power consumed by the relay transmission user m in the D2D communication transmission unit bit video file,
represents the power consumption of the relay transmission user m in the D2D communication transmission video file unit time, Bm,nIndicating the rate at which D2D communication relay user m transmits data to streaming media viewer user n.
Preferably, in step 4, the cooperative reward decision model is configured to determine a reward that the streaming media viewing user needs to pay to the relay transmission user when the cooperative transmission is completed, where the reward is expressed as:
wherein, q (D'n,r) Indicating that the streaming media watching user needs to pay the relay transmission user when the cooperative transmission is completed, bm,n,rIndicating that the relay transmission user m completes the cooperative transmission task D'n,rTransmission cost of bmaxMaximum value of cooperative transmission cost among all relay transmission users, bminRepresents the minimum value of the cooperative transmission cost among all the relay transmission users.
Preferably, the specific implementation manner of step 5 is as follows:
(1) initializing streaming media content requested by each streaming media viewing user according to a reward function decision model, relaying fuzzified collaboration cost consumed by a transmission user, and calculating a collaborative transmission benefit G'm,n,r；
G'm,n,r＝Qn(D'n,r)+Qm(D'n,r)
(2) According to the cooperative transmission effectAnd constructing a cooperative transmission benefit matrix, and initializing a feasible scaling value U of the streaming media watching user and the relay transmission usern,r,Um,r；
Un,r＝max(G'm,n,r),n∈N
Um,r＝0,m∈M
(3) Constructing a weighted bipartite graph according to the cooperative transmission benefit matrix, wherein the weight l is G'm,n,r(ii) a Solving the complete matching of the weighted bipartite graph, and setting the gains corresponding to other code rates of matched streaming media watching users as 0;
(4) if the complete matching of the weighted bipartite graph exists, the complete matching is used as a user matching result in the streaming media cooperative distribution process; if the complete matching of the weighted bipartite graph does not exist, updating the feasible scaling value U of the streaming media watching user and the relay transmission usern,r,Um,rLooping step 3 until the complete match of the weighted bipartite graph is solved; the complete matching of the weighted bipartite graph is a cooperative user matching decision result;
updating feasible scaling values U of streaming media watching users and relay transmission usersn,r,Um,rExpressed as:
U'n,r＝Un,r-min(Un+Um-G'm,n,r),n∈N
U'm,r＝Um,r+min(Un+Um-G'm,n,r),m∈M
wherein, UnFeasible scaling value, U, representing all streaming media watching usersmA feasible scaling value representing all relay transmission users;
(5) based on the rational assumption of users, the benefit constraints of the streaming media watching users and the relay transmission users are set as follows:
Qn(D'n,r)≥qn(D'n,r)
Qm(D'n,r)≥0
wherein q isn(D'n,r) Representing the revenue, Q, of a user watching the streaming media in the direct distribution path of the streaming media contentn(D'n,r) Watch (A)Show for video request D'n,rThe income of the streaming media watching user; qm(D'n,r) Representing D 'for video request'n,rRelaying the user's revenue;
judging whether the matching decisions of the cooperative users respectively meet effect constraints; if so, outputting a code rate selection decision and a relay transmission user matching decision by adopting a streaming media content cooperative distribution path; if not, adopting a direct distribution path of the streaming media content, and determining the code rate selection decision by the bandwidth of the direct distribution of the streaming media.
In another aspect, the present invention provides a mobile streaming media edge collaboration distribution apparatus based on differential privacy, including: the system comprises a data privacy module and a cooperation decision module;
the data privacy module is deployed at a streaming media watching user terminal and a relay transmission user terminal, and the cooperation decision module is deployed at an edge server;
the data privacy module comprises: the system comprises a geographic position privacy sub-module and a cooperation cost privacy sub-module;
the collaboration decision module comprises: a reward function decision sub-module and a cooperative user matching sub-module;
the geographic position privacy submodule is used for collecting the geographic position of the streaming media watching user in each time slice and fuzzifying the geographic position of the streaming media watching user to obtain fuzzified geographic position information;
the cooperation cost privacy sub-module is used for collecting and fuzzifying the cooperation cost of the relay transmission user for completing the streaming media request in each time slice;
the reward function decision submodule is used for calculating benefit values of streaming media content distribution and determining reward which a relay transmission user needs to obtain according to the benefit values and the cooperation cost;
the cooperative user matching sub-module is used for modeling the cooperative distribution process of the streaming media content and making a code rate selection decision and a relay transmission user matching decision for the streaming media distribution according to the reward function decision sub-module;
the mobile streaming media edge collaborative distribution device based on the differential privacy is used for realizing the steps in the mobile streaming media edge collaborative distribution method based on the differential privacy.
One or more technical schemes provided by the invention at least have the following technical effects or advantages:
in the invention, the provided mobile streaming media edge collaboration distribution method and device based on differential privacy can collect the state information of streaming media watching users and relay transmission users in a streaming media edge collaboration framework, including a moving track, a terminal request mode, a D2D communication rate and the like, and provide a localized differential privacy mechanism to effectively protect the security of sensitive data; the method considers D2D communication characteristics to analyze the streaming media data transmission rate, establishes a data privacy model and a cooperation decision model, and establishes an optimization target for maximizing the streaming media content distribution benefit on the premise of meeting data differential privacy and individual rationality constraints; the Kuhn-Munkres algorithm is used for making an optimal streaming media collaborative content distribution user matching mechanism and code rate decision for the edge server, and overall streaming media distribution benefit is improved from the three aspects of data privacy, streaming media watching experience quality and collaborative cost management. The method and the device can prevent the private data of the user from being leaked, improve the experience quality of the streaming media user and reduce the cost of the collaborative distribution of the streaming media.
Drawings
FIG. 1 is a schematic diagram of an apparatus according to an embodiment of the present invention.
Fig. 2 is a flow chart of a method of an embodiment of the present invention.
Fig. 3 is a diagram of multi-user streaming media edge collaboration distribution according to an embodiment of the present invention.
Detailed Description
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
As shown in fig. 1, the embodiment provides a mobile streaming media edge collaborative distribution apparatus based on differential privacy, which includes a data privacy module and a collaborative decision module.
The data privacy module is deployed at the streaming media watching user terminal and the relay transmission user terminal, and is used for collecting and fuzzifying the information of the intelligent terminal user in the mobile edge environment, so that the user data privacy protection is realized, and auxiliary data are provided for the cooperation decision module. The data privacy module comprises two sub-modules: a geographic location privacy sub-module and a collaborative cost privacy sub-module; the geographic position privacy sub-module is responsible for collecting and fuzzifying the geographic position of the streaming media watching user in each time slice; acquiring fuzzified geographic position information at a user terminal through fuzzy parameters distributed by an edge collaboration platform; the collaborative cost privacy submodule is responsible for collecting and obfuscating collaborative cost data that relays the transmission of user-completed streaming media requests at each time slice. The cooperation cost is calculated by the relay transmission user terminal according to the geographical position information, and the fuzzification processing of the data is realized by fuzzification parameters distributed by the edge cooperation platform.
The collaboration decision module is a successor to the data privacy module. The cooperative decision module is deployed in the edge server and can dynamically adjust the cooperative distribution process of the streaming media content according to the video clip information and the fuzzified data collected by the data privacy module. The cooperative decision module comprises a reward function decision submodule and a cooperative user matching submodule; the reward function decision submodule calculates benefit values for streaming media content distribution and determines reward which a relay transmission user needs to obtain according to the benefit values and the cooperative transmission cost; the cooperative user matching sub-module models the cooperative distribution process of the streaming media content, makes a code rate selection decision and a relay transmission user matching decision for the streaming media distribution process by adopting a streaming media cooperative transmission user matching method based on a Kuhn-Munkres algorithm according to the data privacy module and the cooperative decision module, and determines whether the user matching decision meets the revenue constraints of a streaming media watching user and a relay transmission user respectively according to the reward function decision sub-module. If so, outputting a code rate selection decision and a relay transmission user matching decision by adopting a streaming media content cooperative distribution path; if not, adopting a direct distribution path of the streaming media content, and determining the code rate selection decision by the bandwidth of the direct distribution of the streaming media. As shown in fig. 2, an embodiment of the present invention further provides a method for edge collaborative distribution of mobile streaming media based on differential privacy, including the following steps:
step 1, when a streaming media cooperation transmission period begins, collecting resource state information of an edge server and request video clip information of a streaming media watching user.
Namely, the streaming media collaboration distribution device collects the resource status information of the edge server and the content of the video clip requested by the streaming media watching user.
As shown in fig. 3, a streaming media edge collaborative distribution scene is constructed, an edge server is proposed to directly transmit a main link and a collaborative transmission link, and the edge server completes matching of a streaming media viewing user and a relay transmission user. The resource state information of the edge server comprises information such as current network load, calculation resource utilization rate, bandwidth resource allocation condition, channel attenuation parameters and the like; the streaming media request information includes the number of the requested video segment, the length of the requested video segment, and the transmittable bitrate version.
Step 2, acquiring fuzzification request information of a streaming media watching user; and establishing a geographic position privacy model.
Namely, streaming media watching user request information is constructed from a differential privacy mechanism, a geographic position privacy model is established, and differential privacy parameters are determined.
The specific implementation manner of obtaining the fuzzification request information of the streaming media watching user is as follows: constructing a fuzzification function based on the real geographical position of the streaming media watching user and the localized differential privacy to obtain the false geographical position of the streaming media watching user; and acquiring fuzzification request information of the streaming media watching user based on the false geographical position information of the streaming media watching user and the code rate version set of the request video clip.
The real request information of the streaming media watching user is represented as:
Dn,r＝{rn,pn}
wherein D isn,rRepresenting the real request information of the streaming media watching user n in each streaming media cooperation decision period；rnRepresenting a set of bitrate versions of the requested video segment; p is a radical ofnRepresenting the real geographical location of the streaming media viewing user n.
The fuzzification request information of the streaming media watching user is represented as:
D'n,r＝{rn,p'n}
wherein, D'n,rShowing fuzzification request information of a streaming media watching user n in each streaming media cooperation decision period; p'nRepresenting a false geographical location of the streaming media viewing user n. That is, assuming that there are multiple mobile users watching streaming media, it can be represented by the set N ═ {1,2 …, N }. Each streaming media watching user holds an intelligent terminal and is connected to an edge server to request streaming media. The streaming media request of the streaming media watching user n can represent D in each streaming media collaboration decision periodn,r＝{rn,pnIn which r isnRepresenting a set of rate-coded versions, p, of streaming median＝{xn,ynDenotes the user's true geographical location. Fuzzification processing is carried out on the real geographic position in a privacy protection mechanism to obtain p'n＝(x'n,y'n) The obscured streaming media request uploaded by the streaming media viewing user will be denoted as D'n,r＝{Rn,p'n}。
The geographic position privacy model is based on a two-dimensional diagnosis mechanism based on differential privacy, and a formula O (p'n|pn) Can be expressed as:
wherein, O (p'n|pn) A geographic position fuzzification function representing a streaming media viewing user n; p is a radical ofnRepresenting the true geographical location, p, of a streaming media viewing user nn＝{xn,yn}；p'nRepresenting a spurious geographic location, p ', of a streaming media viewing user n'n＝(x'n,y'n) (ii) a Delta represents the information Distortion degree in the two-dimensional Distortion discrimination mechanism and is used for constraining the coordinate deviation of the userA difference; x represents the number of selectable user abscissas and Y represents the number of selectable user ordinates.
Specifically, considering that the edge platform is not trusted, in order to protect the privacy of sensitive data of a streaming media watching user, a translation mechanism needs to be locally adopted in the terminal to obfuscate the geographical position information. The main idea of the distorsion mechanism is to propose a perturbation mechanism based on information Distortion, which is mainly applied in the publishing of local private data. Therefore, the information Distortion degree δ in the discrimination mechanism is defined to constrain the deviation of the abscissa or ordinate of the user, and can be expressed as:
according to the definition of localized differential privacy, epsilon is estimated for a given differential privacy budget1Geographic location obfuscation function O (p ') of streaming media viewing user'n|pn) Need to satisfy epsilon1Differential privacy, i.e.:
if the above equation is satisfied, the information distortion degree delta and the first difference privacy parameter epsilon can be obtained1The constraint relationship of (2).
ε1＝log(XY-1)+2log(1-δ)-log(δ2-2δ)
At this point the differential privacy budget ε is given1The fuzzification function Q (p' | p) based on the exponential mechanism satisfies epsilon1-differential privacy.
Step 3, acquiring the cooperation information of the relay transmission user; and establishing a collaboration cost privacy model.
And constructing the cooperative information of the relay transmission user from the D2D communication principle and a differential privacy mechanism, establishing a cooperative cost privacy model and determining differential privacy parameters.
In each streaming media cooperation decision period of the cooperation information of the relay transmission user, the relay transmission user obtains the rate B of the video file from the edge servermAnd rate R of transmission of video files in D2D communicationm,n. Multiplexing other users' cellular uplinks to transmit video files while using D2D communication, channel gain obeying Rayleigh distribution, rate R of transmitting video files in D2D communicationm,nExpressed as:
wherein B denotes the channel bandwidth, PmAnd PcRespectively representing the terminal transmitting power, N, of a streaming media relay transmission user m and a cellular uplink multiplexing user c0Representing the power of gaussian white noise.
Specifically, in addition to the streaming media viewing user running the streaming media application, it is assumed that there are a plurality of idle users connectable to the edge server, which have the capability of acquiring streaming media resources from the edge server, and at the same time can be connected to the streaming media viewing user through D2D, and it is referred to as streaming media relay transmission user set M ═ 1,2 …, M }. At each streaming media decision period, p is availablem＝{xm,ymRepresents the real position coordinate of the relay transmission user m of the streaming media, BmRepresenting the rate at which video segments are retrieved from the edge server. The streaming media relay transmission user m can communicate with the streaming media watching user n by using D2D, the cellular uplink transmission video files of other users are multiplexed, and the channel gain is subject to Rayleigh distribution. The distance between the streaming media relay transmission user m and the streaming media viewing user n can be represented as dm,nCalculated from the user position coordinates, the channel attenuation coefficient and the Gaussian channel coefficient may be expressed as mu and sigma, respectively0Channel gain σ between streaming media relay transmission user m and streaming media viewing user nm,nCan be expressed as:
the channel gain between the streaming media relay transmission user m and the channel multiplexing user c can be expressed as:
wherein d ism,nRepresents the distance between a streaming media relay transmission user m and a streaming media viewing user n, mu represents the track attenuation coefficient, sigma0Representing the Gaussian channel coefficient, σn,cRepresenting the channel gain between the stream media relay transmission user m and the channel multiplexing user c, dn,cRepresenting the distance between the streaming media viewing user n and the channel multiplexing user c.
After acquiring the streaming media request set, the streaming media relay transmission user requests D 'to any streaming media'n,rGiven a collaborative decision, it can be expressed as
Wherein
Representing selected bitrate of streaming media, bm,nIndicates the bid price, t, for completing the streaming media requestm,nIndicating the time at which the streaming media request is completed. B 'is obtained by fuzzifying real bid price and download time in privacy protection mechanism'm,nAnd t'm,nThus uploaded obfuscated collaborative decisions
The cooperative cost privacy model adopts an exponential mechanism based on differential privacy, and the corresponding cooperative cost fuzzification function P (b' | b) can be expressed as:
and P (b '| b) represents the probability that the real cooperation cost b is converted into any false cooperation cost b', and the fuzzified cooperation cost is output in a random selection mode. q (b' | b) represents a scoring function of the exponential mechanism to measure the score of any spurious cooperation cost, the higher the score, the higher the probability of selecting the spurious cooperation cost as a fuzzy value. Δ b ═ bmax-bminRepresenting the global sensitivity of the scoring function, which is used for restraining the deviation of data fuzzification; epsilon2Representing a second differential privacy parameter.
Specifically, an index mechanism is adopted to perform fuzzification processing on the cooperative transmission cost at the user terminal, so that fuzzified cooperative transmission cost time is obtained. The exponential mechanism is also a mechanism for designing a differential privacy protection algorithm, and the main idea is to select an output value R from an output field R according to the score of a scoring function q, and select the index ratio of the probability of the value and the score. The formal definition of the exponential mechanism can be expressed as: given a scoring function q (D, R) → R, algorithm M satisfies ε -differential privacy if algorithm M satisfies the following equation.
Where Δ q represents the global sensitivity of the scoring function to constrain the bias of data blurring.
To guarantee the effectiveness of the obfuscating function, the scoring function q (b ', b) is a monotonically non-increasing function for the argument | b' -b | and epsilon for a given differential privacy budget2Needs to satisfy epsilon2-differential privacy.
The scoring function q (b', b) is thus set to a linear function:
q(b',b)＝1-|b'-b|
given a differential privacy budget ε2Let b be1And b2Are two non-identical true bid prices, P (b | b)1) And P (b | b)2) Respectively represent the bid price b1And b2The probability of b is formed by fuzzification, and the fuzzification function P (b' | b) based on an exponential mechanism meets the following formula and accords with epsilon2-differential privacy.
And 4, establishing a reward function decision model.
And establishing a reward function decision model according to the geographic position privacy model and the cooperation cost privacy model. The reward function decision model comprises a streaming media viewing benefit model, a relay transmission benefit model and a cooperative reward decision model.
(1) And (5) a streaming media viewing benefit model.
Streaming media watching benefit model income Q of streaming media watching user nnDefined as the difference between the quality of experience gain for viewing streaming media and the cost of paying the collaboration reward, can be expressed as:
wherein Q isn(D'n,r) The method comprises the steps of (1) representing the user benefit of viewing the streaming media, and theta represents the viewing benefit of the streaming media with unit bit rate; h ism(D'n,r) Is a binary variable, h is the matching success between the video watching user n and the relay transmission user mm(D'n,r) 1, otherwise hm(D'n,r)＝0；
Representing the video rate level,/nRepresents the video segment length, sigma represents the economic loss of the viewing streaming media due to the pause,
represents the time consumed by the transmission code rate r from the relay transmission user m to the streaming media watching user nnRepresenting the time when the streaming media watching user n plays the video clip; q. q.sm(D'n,r) The cooperative reward that the streaming media watching user n needs to pay the relay transmission user m is represented; []+Indicates rounding up when
When the temperature of the water is higher than the set temperature,
otherwise
In particular, the profit Q of the user n for viewing the streaming medianDefined as the difference between the quality of experience gain for viewing streaming media and the cost of paying a collaborative reward. Streaming media request D 'for streaming media viewing user n'n,rThe benefit of viewing the streaming media can be expressed as:
wherein, theta represents the viewing income of the streaming media with unit bit rate, sigma represents the incandescence economic loss of viewing the streaming media,
represents the time consumed by the transmission code rate r from the relay transmission user m to the streaming media watching user nnIndicating the time when the streaming media watching user n plays the video segment. When in use
When the temperature of the water is higher than the set temperature,
otherwise
When the cooperative transmission of the streaming media is completed, the streaming media viewing user needs to pay a reward to the cooperative winner as the cost of the cooperative transmission of the streaming media:
request D 'for streaming media'n,rThe benefit of a streaming media viewer n may be denoted as Qn(D'n,r)＝Sn(D'n,r)-Vn(D'n,r)。
(2) A relay transmission benefit model.
The relay transmission benefit model defines the benefit of the streaming media relay transmission user m as the difference between the reward and the cost of the collaborative transmission streaming media can be expressed as:
wherein the content of the first and second substances,
represents the power consumed by the relay transmission user m to download the unit bit video file from the edge terminal,
represents the power consumed by the relay transmission user m in downloading the video file from the edge terminal per unit time,
represents the amount of power consumed by the relay transmission user m in the D2D communication transmission unit bit video file,
indicating the amount of power consumed by the relay transmission user m per unit time for the D2D communication transmission of the video file.
Specifically, the benefit of the streaming media relay transmission user m is defined as the difference between the reward and the cost of the collaborative transmission streaming media. Request D 'for any streaming media'n,rThe reward obtained by the relay transmission user can be expressed as:
Sm(D'n,r)＝qm(D'n,r)
the reward of the relay transmission user m for the cooperative transmission stream media can consist of two parts, on one hand, the stream media relay transmission user m downloads the stream media from the edge server to consume the terminal electricity, and the electricity cost can be expressed as:
wherein the content of the first and second substances,
represents the amount of power consumed by the relay transmission user m to download the unit bit video file from the edge server,
and the power consumption of the relay transmission user m for downloading the video file from the edge server in unit time is shown.
On the other hand, the D2D network transport stream also consumes the terminal power of the streaming media relay user m, and the power cost can be expressed as:
wherein the content of the first and second substances,
represents the amount of power consumed by the relay transmission user m in the D2D communication transmission unit bit video file,
indicating the amount of power consumed by the relay transmission user m per unit time for the D2D communication transmission of the video file.
Thus requesting D 'for streaming media'n,rThe benefit of the streaming media relay transmission user m can be expressed as:
(3) collaborative reward decision model
And the cooperative reward decision model represents that when the cooperative transmission is completed, the streaming media watching user needs to transmit the reward paid by the user to the relay transmission user. Can be expressed as:
wherein, q (D'n,r) Indicating that the streaming media watching user needs to pay the relay transmission user when the cooperative transmission is completed, bm,n,rIndicating that the relay transmission user m completes the cooperative transmission task D'n,rTransmission cost of bmaxMaximum value of cooperative transmission cost among all relay transmission users, bminIndicating the minimum value representing the cooperative transmission cost among all relay transmission users.
And 5, constructing a cooperative user matching model, making a code rate selection decision and a relay transmission user matching decision for the streaming media distribution process by adopting a streaming media cooperative transmission user matching method of a Kuhn-Munkres algorithm according to the established reward function decision model, and determining a streaming media content distribution path.
(1) And matching the model by the collaborative user.
The cooperative user matching model is used for modeling the cooperative distribution process of the streaming media content, the data privacy module and the reward function decision model module make a code rate selection decision and a relay transmission user matching decision for the streaming media distribution process by adopting a streaming media cooperative transmission user matching method based on a Kuhn-Munkres algorithm, and whether the user matching decision meets the profit constraints of a streaming media watching user and a relay transmission user respectively is determined.
The cooperative user matching model comprises the following substeps:
1. initializing streaming media content requested by each streaming media viewing user according to a privacy module and a reward function decision model, relaying and transmitting fuzzified cooperation cost required to be consumed by the users, and calculating cooperation transmission benefit G'm,n,r；G'm,n,r＝Qn(D'n,r)+Qm(D'n,r)；
2. Establishing a cooperative transmission benefit matrix, and initializing a feasible scaling value U of a streaming media watching user and a relay transmission usern,r,Um,r；
Un,r＝max(G'm,n,r),n∈N
Um,r＝0,m∈M
3. Constructing a weighted bipartite graph according to a cooperative transmission benefit matrix, wherein the weight l is G'm,n,r. And solving the complete matching of the bipartite graph, and setting the gains corresponding to other code rates as 0 by the matched streaming media watching user.
4. If the complete matching of the weighted bipartite graph exists, the complete matching is used as a user matching result in the streaming media cooperative distribution process; if not, updating the feasible calibration value U of the streaming media watching user and the relay transmission usern,r,Um,r. And (3) circulating the step until the complete matching of the weighted bipartite graph is solved, wherein the complete matching of the weighted bipartite graph is the matching decision result of the cooperative user.
U'n,r＝Un,r-min(Un+Um-G'm,n,r),n∈N
U'm,r＝Um,r+min(Un+Um-G'm,n,r),m∈M
5. Based on the user rationality assumption, the benefit constraints of the streaming media viewing user and the relay transmission user are expressed as follows:
Qn(D'n,r)≥qn(D'n,r)
Qm(D'n,r)≥0
wherein q isn(D'n,r) Representing the revenue, Q, of a user watching the streaming media in the direct distribution path of the streaming media contentn(D'n,r) Representing D 'for video request'n,rThe income of the streaming media watching user; qm(D'n,r) Representing D 'for video request'n,rAnd relaying the revenue of the user.
Determining whether the user matching decisions respectively satisfy the constraints according to user rational assumptions. If so, outputting a code rate selection decision and a relay transmission user matching decision by adopting a streaming media content cooperative distribution path; if not, adopting a direct distribution path of the streaming media content, and determining the code rate selection decision by the bandwidth of the direct distribution of the streaming media.
And 6, finishing edge cooperative distribution of the streaming media content, and updating the streaming media cooperative transmission period and the state information of an edge server, a streaming media watching user and a relay transmission user.
In conclusion, the mobile streaming media edge collaborative distribution device and method based on the differential privacy, provided by the invention, effectively solve the problem of participating in user data privacy in the streaming media content distribution process in the mobile edge scene, reduce the cost of streaming media collaborative transmission, and improve the experience quality of the streaming media watching user.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to examples, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (10)
1. A mobile streaming media edge cooperative distribution method based on differential privacy is characterized by comprising the following steps:
step 1, when a streaming media cooperation transmission period begins, collecting resource state information of an edge server and request video clip information of a streaming media watching user;
step 2, acquiring fuzzification request information of a streaming media watching user; establishing a geographic position privacy model;
step 3, acquiring the cooperation information of the relay transmission user; establishing a cooperation cost privacy model;
step 4, establishing a reward function decision model, wherein the reward function decision model comprises a streaming media viewing effect model, a relay transmission benefit model and a cooperative reward decision model;
step 5, constructing a cooperative user matching model, making a code rate selection decision and a relay transmission user matching decision for the streaming media distribution process by adopting a streaming media cooperative transmission user matching method based on a Kuhn-Munkres algorithm according to the reward function decision model, and determining a streaming media content distribution path;
and 6, finishing edge cooperative distribution of the streaming media content, and updating the state information of the streaming media cooperative transmission period and the state information of an edge server, a streaming media watching user and a relay transmission user.
2. The edge collaborative distribution method for mobile streaming media based on differential privacy according to claim 1, wherein in step 1, the resource status information of the edge server includes current network load, computational resource usage rate, bandwidth resource allocation condition, and channel weakening parameter;
the request video clip information of the streaming media watching user comprises the number of the request video clip, the length of the request video clip and the code rate version set of the request video clip.
3. The method for edge collaborative distribution of mobile streaming media based on differential privacy as claimed in claim 1, wherein in the step 2, the specific implementation manner for obtaining the obfuscation request information of the streaming media watching user is as follows: constructing a fuzzification function based on the real geographical position of the streaming media watching user and the localized differential privacy to obtain the false geographical position of the streaming media watching user; acquiring fuzzification request information of the streaming media watching user based on the false geographical position information of the streaming media watching user and the code rate version set of the request video clip;
the real request information of the streaming media watching user is represented as:
Dn,r＝{rn,pn}
wherein D isn,rRepresenting the real request information of the streaming media watching user n in each streaming media cooperation decision period; r isnCode representing requested video segmentA set of rate versions; p is a radical ofnRepresenting the real geographical location of the streaming media viewing user n;
the fuzzification request information of the streaming media watching user is represented as:
D'n,r＝{rn,p'n}
wherein, D'n,rShowing fuzzification request information of a streaming media watching user n in each streaming media cooperation decision period; p'nRepresenting a false geographical location of the streaming media viewing user n.
4. The differential privacy-based mobile streaming media edge collaborative distribution method according to claim 1, wherein the geographic location privacy model in the step 2 adopts a two-dimensional warping mechanism based on differential privacy, and a corresponding geographic location fuzzification function is represented as:
wherein, O (p'n|pn) A geographic position fuzzification function representing a streaming media viewing user n; p is a radical ofnRepresenting the true geographical location, p, of a streaming media viewing user nn＝{xn,yn}；p'nRepresenting a spurious geographic location, p ', of a streaming media viewing user n'n＝(x'n,y'n) (ii) a Delta represents the information torsion degree in the two-dimensional torsion mechanism, X represents the number of the selectable user horizontal coordinates, and Y represents the number of the selectable user vertical coordinates;
the information torsion degree and the first differential privacy parameter satisfy the following constraint relationship:
ε1＝log(XY-1)+2log(1-δ)-log(δ2-2δ)
wherein epsilon1Representing a first differential privacy parameter.
5. The method for edge collaborative distribution of mobile streaming media based on differential privacy as claimed in claim 1, wherein in step 3, the relaying of the collaboration information of the user comprises: in each streaming media cooperation decision period, the relay transmission user obtains the rate of the video file from the edge server and the rate of transmitting the video file in D2D communication;
wherein R ism,nRepresents the rate of transmission of video files in D2D communication, B represents the channel bandwidth, PmAnd PcRespectively representing the terminal transmitting power, N, of a streaming media relay transmission user m and a cellular uplink multiplexing user c0Power, σ, representing white Gaussian noisem,nRepresenting the channel gain between a streaming media relay transmission user m and a streaming media viewing user n, dm,nRepresents the distance between a streaming media relay transmission user m and a streaming media viewing user n, mu represents the track attenuation coefficient, sigma0Representing the Gaussian channel coefficient, σn,cRepresenting the channel gain between the stream media relay transmission user m and the channel multiplexing user c, dn,cRepresenting the distance between the streaming media viewing user n and the channel multiplexing user c.
6. The differential privacy-based edge collaborative distribution method for mobile streaming media according to claim 1, wherein in step 3, the collaborative cost privacy model employs an exponential mechanism based on differential privacy, and a corresponding collaborative cost fuzzification function is represented as:
p (b '| b) represents the probability that the real cooperation cost b is converted into any false cooperation cost b', and fuzzified cooperation cost is output in a random selection mode; q (b' | b) represents a scoring function of the exponential mechanism; Δ b ═ bmax-bminRepresenting the global sensitivity of the scoring function; epsilon2Representing a second differential privacy parameter.
7. The method for edge collaborative distribution of mobile streaming media based on differential privacy according to claim 1, wherein in the step 4, the benefit of the streaming media viewing effect model for the streaming media viewing user n is defined as:
wherein Q isn(D'n,r) The method comprises the steps of (1) representing the user benefit of viewing the streaming media, and theta represents the viewing benefit of the streaming media with unit bit rate; h ism(D'n,r) Is a binary variable, h is the matching success between the video watching user n and the relay transmission user mm(D'n,r) 1, otherwise hm(D'n,r)＝0；
Representing the video rate level,/nRepresents the video segment length, sigma represents the economic loss of the viewing streaming media due to the pause,
code rate for transmitting relay transmission user m to stream media watching user n
Time of consumption, τnRepresenting the time, q, at which the streaming media watching user n plays the video segmentm(D'n,r) Agreement indicating that a streaming media watching user n needs to pay a relay transmission user mMaking a reward; []+Indicates rounding up when
When the temperature of the water is higher than the set temperature,
otherwise
In the relay transmission benefit model, the benefit of the streaming media relay transmission user m is defined as:
wherein Q ism(D'n,r) The user benefit of the relay transmission is shown,
represents the amount of power consumed by the relay transmission user m to download the unit bit video file from the edge server,
representing the amount of power consumed by the relay transmitting user m to download the video file from the edge server per unit time, BmIndicating the rate, phi, at which relay transport user m gets video files from the edge server
Represents the amount of power consumed by the relay transmission user m in the D2D communication transmission unit bit video file,
represents the power consumption of the relay transmission user m in the D2D communication transmission video file unit time, Bm,nIndicating the rate at which D2D communication relay user m transmits data to streaming media viewer user n.
8. The method according to claim 7, wherein in step 4, the cooperative reward decision model is used to determine a reward that the streaming media viewing user needs to pay to the relay transmission user when the cooperative transmission is completed, and the reward is expressed as:
wherein, q (D'n,r) Indicating that the streaming media watching user needs to pay the relay transmission user when the cooperative transmission is completed, bm,n,rIndicating that the relay transmission user m completes the cooperative transmission task D'n,rTransmission cost of bmaxMaximum value of cooperative transmission cost among all relay transmission users, bminRepresents the minimum value of the cooperative transmission cost among all the relay transmission users.
9. The mobile streaming media edge collaborative distribution method based on differential privacy according to claim 1, wherein the specific implementation manner of the step 5 is as follows:
(1) initializing streaming media content requested by each streaming media viewing user according to a reward function decision model, relaying fuzzified collaboration cost consumed by a transmission user, and calculating a collaborative transmission benefit G'm,n,r；
G'm,n,r＝Qn(D'n,r)+Qm(D'n,r)
(2) According to the cooperative transmission benefit, a cooperative transmission benefit matrix is constructed, and a feasible scaling value U of a streaming media watching user and a relay transmission user is initializedn,r,Um,r；
Un,r＝max(G'm,n,r),n∈N
Um,r＝0,m∈M
(3) Constructing a weighted bipartite graph according to the cooperative transmission benefit matrix, wherein the weight is lG'm,n,r(ii) a Solving the complete matching of the weighted bipartite graph, and setting the gains corresponding to other code rates of matched streaming media watching users as 0;
(4) if the complete matching of the weighted bipartite graph exists, the complete matching is used as a user matching result in the streaming media cooperative distribution process; if the complete matching of the weighted bipartite graph does not exist, updating the feasible scaling value U of the streaming media watching user and the relay transmission usern,r,Um,rLooping step 3 until the complete match of the weighted bipartite graph is solved; the complete matching of the weighted bipartite graph is a cooperative user matching decision result;
updating feasible scaling values U of streaming media watching users and relay transmission usersn,r,Um,rExpressed as:
U'n,r＝Un,r-min(Un+Um-G'm,n,r),n∈N
U'm,r＝Um,r+min(Un+Um-G'm,n,r),m∈M
wherein, UnFeasible scaling value, U, representing all streaming media watching usersmA feasible scaling value representing all relay transmission users;
(5) based on the rational assumption of users, the benefit constraints of the streaming media watching users and the relay transmission users are set as follows:
Qn(D'n,r)≥qn(D'n,r)
Qm(D'n,r)≥0
wherein q isn(D'n,r) Representing the revenue, Q, of a user watching the streaming media in the direct distribution path of the streaming media contentn(D'n,r) Representing D 'for video request'n,rThe income of the streaming media watching user; qm(D'n,r) Representing D 'for video request'n,rRelaying the user's revenue;
judging whether the matching decisions of the cooperative users respectively meet effect constraints; if so, outputting a code rate selection decision and a relay transmission user matching decision by adopting a streaming media content cooperative distribution path; if not, adopting a direct distribution path of the streaming media content, and determining the code rate selection decision by the bandwidth of the direct distribution of the streaming media.
10. A mobile streaming media edge collaboration distribution apparatus based on differential privacy, comprising: the system comprises a data privacy module and a cooperation decision module;
the data privacy module is deployed at a streaming media watching user terminal and a relay transmission user terminal, and the cooperation decision module is deployed at an edge server;
the data privacy module comprises: the system comprises a geographic position privacy sub-module and a cooperation cost privacy sub-module;
the collaboration decision module comprises: a reward function decision sub-module and a cooperative user matching sub-module;
the geographic position privacy submodule is used for collecting the geographic position of the streaming media watching user in each time slice and fuzzifying the geographic position of the streaming media watching user to obtain fuzzified geographic position information;
the cooperation cost privacy sub-module is used for collecting and fuzzifying the cooperation cost of the relay transmission user for completing the streaming media request in each time slice;
the reward function decision submodule is used for calculating benefit values of streaming media content distribution and determining reward which a relay transmission user needs to obtain according to the benefit values and the cooperation cost;
the cooperative user matching sub-module is used for modeling the cooperative distribution process of the streaming media content and making a code rate selection decision and a relay transmission user matching decision for the streaming media distribution according to the reward function decision sub-module;
the differential privacy-based mobile streaming media edge collaborative distribution apparatus is used for implementing the steps in the differential privacy-based mobile streaming media edge collaborative distribution method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011140075.1A CN112423027B (en) | 2020-10-22 | 2020-10-22 | Mobile streaming media edge collaboration distribution device and method based on differential privacy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011140075.1A CN112423027B (en) | 2020-10-22 | 2020-10-22 | Mobile streaming media edge collaboration distribution device and method based on differential privacy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112423027A true CN112423027A (en) | 2021-02-26 |
| CN112423027B CN112423027B (en) | 2021-10-22 |
Family
ID=74841647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011140075.1A Active CN112423027B (en) | 2020-10-22 | 2020-10-22 | Mobile streaming media edge collaboration distribution device and method based on differential privacy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112423027B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113727358A (en) * | 2021-08-31 | 2021-11-30 | 河北工程大学 | KM and greedy algorithm-based edge server deployment and content caching method |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120042390A1 (en) * | 2010-08-12 | 2012-02-16 | Nokia Corporation | Method and apparatus for secure revocable location sharing |
| CN103476140A (en) * | 2013-09-05 | 2013-12-25 | 北京邮电大学 | Method for making choice between D2D cooperating multicast mode and fixed relay cooperating multicast mode in cellular system |
| CN107707530A (en) * | 2017-09-12 | 2018-02-16 | 福建师范大学 | A kind of method for secret protection and system of mobile intelligent perception |
| EP3420465A1 (en) * | 2016-02-25 | 2019-01-02 | ACS (US), Inc. | Platform for computing at the mobile edge |
| US10289816B1 (en) * | 2018-06-08 | 2019-05-14 | Gsfm Llc | Methods, systems, and devices for an encrypted and obfuscated algorithm in a computing environment |
| CN110069940A (en) * | 2019-03-15 | 2019-07-30 | 上海大学 | Intimacy protection system and method in horde sensing network based on incentive mechanism |
| CN110213627A (en) * | 2019-04-23 | 2019-09-06 | 武汉理工大学 | Flow medium buffer distributor and its working method based on multiple cell user mobility |
| CN110611667A (en) * | 2019-09-02 | 2019-12-24 | 深圳市赛梅斯凯科技有限公司 | Dynamic position privacy protection method and device in edge computing environment |
| US20200082442A1 (en) * | 2017-09-19 | 2020-03-12 | Leonard Z. Sotomayor | Systems apparatus and methods for management and distribution of video content |
| CN111431630A (en) * | 2020-05-25 | 2020-07-17 | 河海大学常州校区 | AUV (autonomous underwater vehicle) cooperation source node position privacy protection method based on anonymous cluster in UASNs (Universal asynchronous receiver network) |
| CN111723402A (en) * | 2020-06-21 | 2020-09-29 | 天津理工大学 | Flow compensation excitation method facing MDU privacy data protection based on QL learning strategy |
-
2020
- 2020-10-22 CN CN202011140075.1A patent/CN112423027B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120042390A1 (en) * | 2010-08-12 | 2012-02-16 | Nokia Corporation | Method and apparatus for secure revocable location sharing |
| CN103476140A (en) * | 2013-09-05 | 2013-12-25 | 北京邮电大学 | Method for making choice between D2D cooperating multicast mode and fixed relay cooperating multicast mode in cellular system |
| EP3420465A1 (en) * | 2016-02-25 | 2019-01-02 | ACS (US), Inc. | Platform for computing at the mobile edge |
| CN107707530A (en) * | 2017-09-12 | 2018-02-16 | 福建师范大学 | A kind of method for secret protection and system of mobile intelligent perception |
| US20200082442A1 (en) * | 2017-09-19 | 2020-03-12 | Leonard Z. Sotomayor | Systems apparatus and methods for management and distribution of video content |
| US10289816B1 (en) * | 2018-06-08 | 2019-05-14 | Gsfm Llc | Methods, systems, and devices for an encrypted and obfuscated algorithm in a computing environment |
| CN110069940A (en) * | 2019-03-15 | 2019-07-30 | 上海大学 | Intimacy protection system and method in horde sensing network based on incentive mechanism |
| CN110213627A (en) * | 2019-04-23 | 2019-09-06 | 武汉理工大学 | Flow medium buffer distributor and its working method based on multiple cell user mobility |
| CN110611667A (en) * | 2019-09-02 | 2019-12-24 | 深圳市赛梅斯凯科技有限公司 | Dynamic position privacy protection method and device in edge computing environment |
| CN111431630A (en) * | 2020-05-25 | 2020-07-17 | 河海大学常州校区 | AUV (autonomous underwater vehicle) cooperation source node position privacy protection method based on anonymous cluster in UASNs (Universal asynchronous receiver network) |
| CN111723402A (en) * | 2020-06-21 | 2020-09-29 | 天津理工大学 | Flow compensation excitation method facing MDU privacy data protection based on QL learning strategy |
Non-Patent Citations (4)
| Title |
|---|
| QIUCHENG MIA: "Differential privacy–based location privacy enhancing in edge computing", 《CONCURRENCY AND COMPUTATION: PRACTICE AND EXPERIENCE》 * |
| YONGFENG QIAN: "Privacy-preserving based task allocation with mobile edge clouds", 《INFORMATION SCIENCES》 * |
| 张鹏程: "一种移动边缘环境下面向隐私保护QoS预测方法", 《计算机学报》 * |
| 钱志鸿: "面向5G通信网的D2D技术综述", 《通信学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113727358A (en) * | 2021-08-31 | 2021-11-30 | 河北工程大学 | KM and greedy algorithm-based edge server deployment and content caching method |
| CN113727358B (en) * | 2021-08-31 | 2023-09-15 | 河北工程大学 | Edge server deployment and content caching method based on KM and greedy algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112423027B (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Game theory and reinforcement learning based secure edge caching in mobile social networks | |
| Luo et al. | Adaptive video streaming with edge caching and video transcoding over software-defined mobile networks: A deep reinforcement learning approach | |
| He et al. | Secure social networks in 5G systems with mobile edge computing, caching, and device-to-device communications | |
| CN108990159B (en) | Combined resource allocation method based on hierarchical game in mobile edge computing system | |
| CN111132077B (en) | Multi-access edge computing task unloading method based on D2D in Internet of vehicles environment | |
| Kazmi et al. | Infotainment enabled smart cars: A joint communication, caching, and computation approach | |
| Zhou et al. | Communication-and-computing latency minimization for UAV-enabled virtual reality delivery systems | |
| Jameel et al. | Performance analysis of cooperative V2V and V2I communications under correlated fading | |
| US20140243009A1 (en) | Whitespace channel allocation | |
| Wang et al. | Price the QoE, not the data: SMP-economic resource allocation in wireless multimedia Internet of Things | |
| Kim | 5G Network Communication, Caching, and Computing Algorithms Based on the Two‐Tier Game Model | |
| Chen et al. | Cross-layer QoE optimization for D2D communication in CR-enabled heterogeneous cellular networks | |
| Chen et al. | Collaborative content placement among wireless edge caching stations with time-to-live cache | |
| Liu et al. | Resource allocation for video transcoding and delivery based on mobile edge computing and blockchain | |
| CN112423027B (en) | Mobile streaming media edge collaboration distribution device and method based on differential privacy | |
| Ye et al. | Power control for wireless VBR video streaming: From optimization to reinforcement learning | |
| Baccour et al. | Collaborative hierarchical caching and transcoding in edge network with CE-D2D communication | |
| Belmoukadam et al. | On accounting for screen resolution in adaptive video streaming: QoE‐driven bandwidth sharing framework | |
| Zhang et al. | A signaling-based incentive mechanism for device-to-device content sharing in cellular networks | |
| Hosseini et al. | SVC-based multi-user streamloading for wireless networks | |
| Razzac et al. | Dimensioning and profit sharing in hybrid LTE/DVB systems to offer mobile TV services | |
| CN112887314B (en) | Time delay perception cloud and mist cooperative video distribution method | |
| Li et al. | Transmission mode selection and interference mitigation for social aware D2D communication | |
| Park et al. | Layer-assisted video quality adaptation for improving QoE in wireless networks | |
| Yin et al. | Joint cloud computing and wireless networks operations: a game theoretic approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |